Methods for characterizing colour scanners and digital cameras

Abstract

After a brief introduction to the principles of colour management of both device dependent and independent systems the principles of scanner and camera characterisation are discussed in relation to the ANSI/ISO tools developed for this purpose. Procedures are outlined for the evaluation of such devices in terms of their ability to provide satisfactory colour reproduction, and then the techniques which may be used to achieve the necessary characterisation are discussed. These include least-squares fitting, non-linear transformations, dye combination modelling and look-up-tables with interpolation. Many of the important issues associated with characterisation, such as gamut mapping and appearance modelling, are not discussed in this paper but in the paper on printer characterisation also in this issue.

The chapter published was an update of an earlier paper for a special issue of Displays Journal, published by SID, dealing with colour management. The written paper was based on industrial research and development that Tony Johnson had carried out over a number of years, and some of his work in helping to develop International Standards in Graphic Technology. It described possible solutions to the problem of unambiguously measuring the colour of image pixels, when capturing them digitally, so that they can be accurately reproduced. This problem was a significant one in the process of colour management that was evolving at that time and one which Tony Johnson was particularly qualified to discuss because of his industrial research experience of colour imaging over the preceding 20 years. The paper had originally been written when colour management techniques were still being developed but by 2002 was becoming well established. The techniques discussed in that paper are now widely used.
The paper described the background to the problem, techniques for characterising a device (i.e. developing a mathematical transformation that enables the device signals to be converted to an internationally recognized measurement system) using standard test images developed by ISO TC130, and the form of various mathematical transformations that can be provided by such targets.

• http://www.sciencedirect.com/science?_ob...